| name | briefing |
| description | Produce a "catch me up" briefing — a synthesized start-of-day / overnight summary across email, calendar, tasks, and chat, ending in suggested next actions. Use when the user says "catch me up", "what did I miss", "morning briefing", "what's my day", "start of day". Not for acting on findings (sending, scheduling, creating tasks) — it surfaces and suggests; you approve. Not for single-source lookups or deep person/meeting prep. |
Catch-Me-Up Briefing
Synthesize one prioritized brief across the user's inboxes, calendar, tasks, and
chat. The job is surface and suggest — present what needs attention and a
ranked list of suggested actions; the user decides. Never send, schedule, or
mutate anything without approval.
This skill is the runnable procedure; the briefing way carries when to reach
for it and when a single-source query is enough instead.
Tool-agnostic by design
Map the scout domains below to whatever accounts and services are actually
connected this session — email, calendar, task, and chat tools vary per setup, so
discover them rather than assuming a provider. If a domain has no connected tool,
say so in the brief rather than skipping it silently.
This skill deliberately sets no allowed-tools: it must reach whatever EA
tools are connected and spawn scout subagents, which can't be enumerated ahead of
time. Its tight description is what keeps it in its lane.
Gather in parallel
For a multi-account, multi-service briefing, spawn parallel scout subagents so the
slow I/O overlaps — and so each scout's raw fetch stays out of the lead's context:
| Scout | Domain | Shares with |
|---|
| inbox-scout | email (all accounts) — fetch, classify, filter | people/topics → ops-scout |
| ops-scout | calendar + tasks — schedule, status, cross-ref vs. inbox | the lead |
| + specialists | chat, meeting recaps, files — as connected | the lead |
Keep it to 3–5 scouts. Scouts share cross-reference-worthy findings as they go;
each returns one structured report; the lead waits for all reports before
synthesizing. Skip the team for a single account/service or a quick query —
just gather directly (that judgment lives in the briefing way).
Synthesize (priority order)
- Overdue — past-due tasks, priority-ranked; broken commitments first.
- Today's schedule — events, each with related task count + email context.
- Due today — tasks due today, not yet complete.
- Action required — emails awaiting reply, urgency-ranked; note cross-refs.
- Already addressed — replied items; flag any that resolve open tasks.
- Open tasks in play — pending tasks linked to today's mail/calendar.
- Eisenhower snapshot — one line (X do-first, Y to schedule, Z in inbox).
- Cross-references — key connections found across mail, tasks, calendar.
- Suggested actions — what to tackle first, given calendar constraints.
Then: suggested task mutations
From the cross-references, surface a numbered list the user can approve, modify,
or dismiss — never applied automatically:
- Create — obligations from email not yet tracked.
- Complete — tasks the sent mail appears to resolve.
- Update — tasks whose priority/due date shifted.
- Stale? — tasks not referenced in recent comms.
Not for
- Acting on findings — sending replies, scheduling, creating/closing tasks. It
surfaces and suggests; the user approves each.
- Single-source lookups — just query that inbox/calendar directly.
- Deep person/topic/meeting preparation — that's the intelligence way's
cross-referencing research, a separate workflow.
See also
- the briefing way — when to brief vs. when a single query suffices
- the ea / email / calendar / tasks / comms ways — per-domain judgment